Deleted
Deleted Member
Posts: 0
Karma:
|
Post by Deleted on Dec 7, 2016 21:27:55 GMT -5
I am going to go over this again. Under a Elo type matchmaker that Pixo is planning on, if you win most of the time (which I am assuming that you do), your ranking would go up and you will keep facing progressively higher ranked players until you no longer win most of the time, and you can bet your soul that the large majority players you end up with are going to have much better hangars than you. You can say goodbye to relatively fair matches (i.e the ones you can carry) because the new MM will try to make your matches "fair" by giving your opponents a material one (i.e. MUCH BETTER hangars) to offset your skill advantage. The only players that stand to gain from the new MM are players who have no idea what they are doing and maxed hangars, since they cannot be matched any higher regardless of how well they actually do. This. Ever since Pix talked about basing the new matchmaker on "player achievements" I have been uneasy. Discriminating based on success rather than the composition of your hangar makes it much simpler for the devs, but it definitely comes at a price. Basically you want to go towards an end-game type hangar asap in this scenario. Prior to that, if you have a run of success you will simply be beaten back into your proper place by the progressively steeper material edge being given to your enemies. I would see "fair" matchmaking as forming a pool of players who have similar material assets, then use perceived skill / success to balance the two sides. Apparently Pix thinks otherwise.
|
|
|
Post by ShutUpAndSmokeMyWeed on Dec 7, 2016 21:34:33 GMT -5
I am going to go over this again. Under a Elo type matchmaker that Pixo is planning on, if you win most of the time (which I am assuming that you do), your ranking would go up and you will keep facing progressively higher ranked players until you no longer win most of the time, and you can bet your soul that the large majority players you end up with are going to have much better hangars than you. You can say goodbye to relatively fair matches (i.e the ones you can carry) because the new MM will try to make your matches "fair" by giving your opponents a material one (i.e. MUCH BETTER hangars) to offset your skill advantage. The only players that stand to gain from the new MM are players who have no idea what they are doing and maxed hangars, since they cannot be matched any higher regardless of how well they actually do. This. Ever since Pix talked about basing the new matchmaker on "player achievements" I have been uneasy. Discriminating based on success rather than the composition of your hangar makes it much simpler for the devs, but it definitely comes at a price. Basically you want to go towards an end-game type hangar asap in this scenario. Prior to that, if you have a run of success you will simply be beaten back into your proper place by the progressively steeper material edge being given to your enemies. I would see "fair" matchmaking as forming a pool of players who have similar material assets, then use perceived skill / success to balance the two sides. Apparently Pix thinks otherwise. But what is the alternative to rewarding people for progressing to the highest tier in which they're competitive? You get sealclubbing... Also you must keep in mind that fast matchmaking is a top priority for casual games like this. Yes, it would be ideal to have even skills and hangars, but not practical on this platform.
|
|
Deleted
Deleted Member
Posts: 0
Karma:
|
Post by Deleted on Dec 7, 2016 21:52:39 GMT -5
Most of sealclubbing comes from the fact that Pix deliberately chooses to underweight bots like the Gepard in the matchmaker algorithm. I suspect this came from wanting to reward those who spend $$$ to get Au and premium bots... but the consequences have been unfortunate.
Pix could use the vast amount of data they've collected to model the effectiveness of every bot and weapon and use those profiles to create a high degree of material parity in matches. They simply choose not to do so. They'd rather keep dropping crap players with heavy hangars in the rankings until they can get to a place where they can win through sheer health/firepower rather than actually learning to play well with the stuff they have. Hence the proliferation of 6/5 heavies in what used to be mid-medium territory recently. If they go any further down this path, expect more of that kind of thing.
|
|
Deleted
Deleted Member
Posts: 0
Karma:
|
Post by Deleted on Dec 8, 2016 0:42:25 GMT -5
I am going to go over this again. Under a Elo type matchmaker that Pixo is planning on, if you win most of the time (which I am assuming that you do), your ranking would go up and you will keep facing progressively higher ranked players until you no longer win most of the time, and you can bet your soul that the large majority players you end up with are going to have much better hangars than you. You can say goodbye to relatively fair matches (i.e the ones you can carry) because the new MM will try to make your matches "fair" by giving your opponents a material one (i.e. MUCH BETTER hangars) to offset your skill advantage. The only players that stand to gain from the new MM are players who have no idea what they are doing and maxed hangars, since they cannot be matched any higher regardless of how well they actually do. This. Ever since Pix talked about basing the new matchmaker on "player achievements" I have been uneasy. Discriminating based on success rather than the composition of your hangar makes it much simpler for the devs, but it definitely comes at a price. Basically you want to go towards an end-game type hangar asap in this scenario. Prior to that, if you have a run of success you will simply be beaten back into your proper place by the progressively steeper material edge being given to your enemies. I would see "fair" matchmaking as forming a pool of players who have similar material assets, then use perceived skill / success to balance the two sides. Apparently Pix thinks otherwise. Great post. You and Curry Pot have expressed the crux of this in the most effective manner for my understanding at least. So for those of us who are good pilots (or at least decent, like me, but that's even stretching it) who are grinding their way into a good gold or top tier hangar, but are in no way close yet, we shall suffer the material difference of this new MM in an even more harsh and excruciating way than before? Sweet. Guess my plan to spam some freakin Ag for upgrades is going to proceed apace then.
|
|
|
Post by CΛΜΡΞΓ™ on Dec 8, 2016 0:50:43 GMT -5
Great post. You and Curry Pot have expressed the crux of this in the most effective manner for my understanding at least. So for those of us who are good pilots (or at least decent, like me, but that's even stretching it) who are grinding their way into a good gold or top tier hangar, but are in no way close yet, we shall suffer the material difference of this new MM in an even more harsh and excruciating way than before? Sweet. Guess my plan to spam some freakin Ag for upgrades is going to proceed apace then. honestly as it is now you face gepard waves in bronze/silver and overall terrible MM in gold. I don't see how it could be much worse. im sure they will Find a way to make it fair.
|
|
|
Post by KaneoheGrown on Dec 8, 2016 1:12:25 GMT -5
The fact that players can game the system to their advantage already speaks to the huge imbalance currently in the system. Simply matching hangars based on weapons and bots has already been shown to be a sub par method for matchmaking. An ELO system allows hangar make up to factor into matchmaking WITH player skill, something that isn't taken into account for currently. The arguments against this seem to hinge greatly on excluding additional data rather than add more. To ensure balanced matches, MORE data should be applied to the underlying matchmaking algorithm, not less.
As it stands, someone who unknowingly levels a wep or bot beyond the rest of their hangar is thrown into matches where they're outgunned. Losses incurred from that mistake will allow mm to drop that player to a more suitable mm pool.
In turn a player who is highly skilled but with a sub par hangar will play against better equipped/lower skilled players. If they win they'll move up till their playing in matches where their higher skill/poor equipment will be balanced.
In the end, enhancing the current mm system is better for the entire platform. The biggest advantage being it keeps players equally matched in BOTH skill & equipment.
|
|
|
Post by CΛΜΡΞΓ™ on Dec 8, 2016 1:14:33 GMT -5
The fact that players can game the system to their advantage already speaks to the huge imbalance currently in the system. Simply matching hangars based on weapons and bots has already been shown to be a sub par method. An ELO system allows hangar make up to factor into matchmaking WITH player skill, something that isn't taken into account for currently. The arguments against this seem to hinge greatly on excluding additional data rather than add more. To ensure balanced matches, MORE data should be applied to the underlying matchmaking algorithm, not less. As it stands, someone who unknowingly levels a wep or bot beyond the rest of their hangar is thrown into matches where they're outgunned. Losses incurred from that mistake will allow mm to drop that player to a more suitable mm pool. In turn a player who is highly skilled but with a sub par hangar will play against better equipped/lower skilled players. If they win they'll move up till their playing in matches where there higher skill/poor equipment will be balanced. In the end, enhancing the current mm system is better for the entire platform. The biggest advantage being it keeps players equally matched in BOTH skill & equipment. this sounds good and all, but have you ever played a low gold hangar in a top tier match? No matter how skilled you are with your bots, you still feel like a burden and it is not a fun experience.
|
|
|
Post by Strayed on Dec 8, 2016 1:17:54 GMT -5
|
|
|
Post by KaneoheGrown on Dec 8, 2016 1:20:50 GMT -5
Actually yes I did. There was a good two month period (around June I believe) mm was screwed up. My 6/9 hangar was regularly pitted to top gold hangars.
Sure it wasn't always fun, but an ELO system would have dropped me when my win rate was dropped significantly.
The problem currently is that while I was in that limbo period there was nothing to trigger bumping me down.
This is precisely why we need player skill to be factored into MM.
Honestly, the AVERAGE win rate of all players should be around 50% from a statistical standpoint. ELO would ensure that that's what happens. The only players outside of this range would be the best players with maxed hangars as MM wouldn't have anyone "higher" to pit them against.
|
|
|
Post by KaneoheGrown on Dec 8, 2016 1:27:38 GMT -5
Additionally, Strayed has a point. "Git Gud" would apply, meaning players are no longer incentivized to create a "maxed for tier" hangar. Either continue upgrading, or be forced into tougher matches.
Seal clubbing is seal clubbing. Whether a 5 slot 4/12 Gep hangar, or a maxed high silver hangar. ELO removes this ability and ensures that players continuously aim for top tier. The "I'm just 'farming' for gold tier" excuse will dry up pretty quick when you're not able to exploit lower level/equipped players, thus ensuring a fluid growth of top tier play.
|
|
|
Post by Strayed on Dec 8, 2016 1:31:35 GMT -5
You'll see in the match above that I posted(It is fairly old though), my stats were comparable to and even sometimes exceeding that of the other players even though my bots and weaps were 3 levels lower on average. The question is, should I have been placed up against comparable opponents or should I have been allowed to utterly wreck anybody coming into low gold?
|
|
Deleted
Deleted Member
Posts: 0
Karma:
|
Post by Deleted on Dec 8, 2016 1:40:43 GMT -5
You'll see in the match above that I posted(It is fairly old though), my stats were comparable to and even sometimes exceeding that of the other players even though my bots and weaps were 3 levels lower on average. The question is, should I have been placed up against comparable opponents or should I have been allowed to utterly wreck anybody coming into low gold? Admittedly that's impressive. But but but...the enemy team had a natasha and thunder raijin. Seriously, a gekko natasha. ? I kid. Well done.
|
|
|
Post by KaneoheGrown on Dec 8, 2016 1:53:36 GMT -5
You'll see in the match above that I posted(It is fairly old though), my stats were comparable to and even sometimes exceeding that of the other players even though my bots and weaps were 3 levels lower on average. The question is, should I have been placed up against comparable opponents or should I have been allowed to utterly wreck anybody coming into low gold? Agreed, perfect example of how a hangar pushed "up" is still competitive. We see it all the time on the forum about how "skill" matters (which we know clearly does). The fact that MM currently isn't factoring that in is a detriment, not an advantage to the game. I played Clash Royale and Clash of Clans, both tier systems were beneficial (overall). While players could sandbag in those games, it's much more difficult to do that in a game where resource generation isn't time based, but activity based. Unlike CoC the only way to generate Au or Ag in War Robots is to earn it through match playing. So while some would bemoan (theoretical) sandbagging, the current game structure simply won't allow it over the long term. Tie in rewards "based" on rank/tier level of play, you even further remove incentives to drop one's rank. Ex: Say a maxed hangar sandbags to low silver ranking. If MM only rewards 3 Au (max) and 1/4 the Ag (say 100k) compared to a top tier match (10 Au, 500k+ Ag), then the player wastes far more of their playing time for meager rewards. Thus an incentive for them to remain in top end play. (*Note: the rewards I suggested were pulled out of thin air for my example)
|
|
|
Post by CΛΜΡΞΓ™ on Dec 8, 2016 2:02:20 GMT -5
It's really not a good way to match in gold though. Deathbutton setups (especially rhino) do not perform their role properly, as 1 salvo is not enough to kill even a Griffin. enemy midrange becomes 'take off half or more of your bots hp in 1 salvo', pretty much crippling your bots for good instead of simply puncturing them. you can say it's a matter of getting good, but really it's a completely different game trying to fit in with underlevelled bots.
|
|
|
Post by Zhøu™ on Dec 8, 2016 4:47:03 GMT -5
You'll see in the match above that I posted(It is fairly old though), my stats were comparable to and even sometimes exceeding that of the other players even though my bots and weaps were 3 levels lower on average. The question is, should I have been placed up against comparable opponents or should I have been allowed to utterly wreck anybody coming into low gold? Agreed, perfect example of how a hangar pushed "up" is still competitive. We see it all the time on the forum about how "skill" matters (which we know clearly does). The fact that MM currently isn't factoring that in is a detriment, not an advantage to the game. I played Clash Royale and Clash of Clans, both tier systems were beneficial (overall). While players could sandbag in those games, it's much more difficult to do that in a game where resource generation isn't time based, but activity based. Unlike CoC the only way to generate Au or Ag in War Robots is to earn it through match playing. So while some would bemoan (theoretical) sandbagging, the current game structure simply won't allow it over the long term. Tie in rewards "based" on rank/tier level of play, you even further remove incentives to drop one's rank. Ex: Say a maxed hangar sandbags to low silver ranking. If MM only rewards 3 Au (max) and 1/4 the Ag (say 100k) compared to a top tier match (10 Au, 500k+ Ag), then the player wastes far more of their playing time for meager rewards. Thus an incentive for them to remain in top end play. (*Note: the rewards I suggested were pulled out of thin air for my example) Speaking of CR, their trophy system pisses me off really badly. I had to quit because I couldn't keep my emotions in check. Almost hit 3k before Sparky, Miner and Lava Hound hit....
|
|
|
Post by Muhlakai on Dec 8, 2016 9:25:04 GMT -5
Pix could use the vast amount of data they've collected to model the effectiveness of every bot and weapon and use those profiles to create a high degree of material parity in matches. They simply choose not to do so. If they go any further down this path, expect more of that kind of thing. Absolutely. I posited in another thread that every week they could process the win % for each bot/weapon (incl. levels) and adjust MM scores dynamically and automatically. Then based on those scores a player skill score could be determined based on whether a player won more or less than expected with the equipment they used. This could be averaged even from each and every match in the week.
|
|
|
Post by SlowReflexes on Dec 8, 2016 9:55:35 GMT -5
Be careful of what you wish for. You just might get it... Good and hard.
|
|
|
Post by lephturn on Dec 8, 2016 10:00:18 GMT -5
There is a balance to strike here. Certainly they don't have that balance right yet. As a brand new player who came to this board 1 day after joining and spent $5 on a Gepard with 3 Pinatas at level 5 I have a ridiculous win rate.
That said, taking player skill into account is important, but if you go too far you end up with everyone close to 50% War Robots and it could be frustrating for good players who's reward for getting better is to be placed against better opponents. Maybe that's not such a bad thing... and to deal with that you swap the focus on win rate to a focus on rank.
The fact that it is possible for me to join and within 3 days have a win rate over 70% says the current system is completely broken so I'm happy to hear they are going to do something about it.
There are some simple things they could do in-game - like making the recommendations actually helpful. Limit weapon levels to within 2 or 3 plus or minus of bot level. Don't offer medium and heavy bots until the player has powerful enough lights. Guide players to build a balanced hangar. It's not all in the MM...
|
|
|
Post by SlowReflexes on Dec 8, 2016 10:13:20 GMT -5
I don't think people are grasping how fundamentally different the new system is going to be.
In our current system it more or less segregates people by matching like equipment against like, and the measure of how good you are is your win %. (Lots of ways to game that, thus clubbing, but that's the essence. Ties in well with the monetary in game rewards system, which depends on wins.)
The new system will explicitly adjust matches so that everyone approaches a 50% win rate. The variable here, will be the 'quality' of your opposition- the better you are the higher quality the opposition. In this case, the measure of how good you are will be your ranking, which will be some kind of representation of your strength as a player. (With you being matched against players of eeual ranking, and how everyone does will adjust their ranking, of course within constraints of 50% win rates for all.)
That's nice for bragging rights, but doesn't fit *at all* with the current in game monetary reward system. And seeing as how upgrades in this game are critical and *will* directly bear on your strength.... Something has to give. Among the other complaints I've expressed about the model behind the proposed new MM system, as the game currently stands this will explicitly make this game PtW.
|
|
|
Post by SlowReflexes on Dec 8, 2016 10:17:58 GMT -5
Also, the new system will provide incentives against ever using anything but your most powerful hangar. Think about what that will do to sales of expensive equipment that does not currently scale up to higher tier play. You want Pix to go broke?
|
|
|
Post by ⓣⓡⓘⓒⓚⓨ48 on Dec 8, 2016 10:45:41 GMT -5
Just a thought... But I noticed that we are making predictions of doom and gloom for a MM system that hasn't even been finished yet, from bits and pieces we have heard. We don't know all the tweaks and quirks that they will put in it by the time it is released, so prophesying the worst case scenarios may not be the best way to look at it. It's good to point out the possible pitfalls... but not to expect it to suck before we even know all the details yet. IMHO. Yes, I agree the possibilities will exist for the worst case to happen... but until we actually know what they are doing... getting all grumpy about what may happen is going to cause you undo stress. Might need to find happybobross and do a painting with him to help put things into perspective Either way, I am glad they are looking at changing it... no doubt, at all, that it is needed.
|
|
|
Post by ⓣⓡⓘⓒⓚⓨ48 on Dec 8, 2016 10:46:52 GMT -5
BTW... that game is impressive Strayed.
|
|
|
Post by Muhlakai on Dec 8, 2016 10:48:32 GMT -5
In our current system it more or less segregates people by matching like equipment against like, and the measure of how good you are is your win %. The new system will explicitly adjust matches so that everyone approaches a 50% win rate. In this case, the measure of how good you are will be your ranking, which will be some kind of representation of your strength as a player. That's nice for bragging rights, but doesn't fit *at all* with the current in game monetary reward system. As the game currently stands this will explicitly make this game PtW. Not at all. You're going to have stiffer competition for that Au if your teammates are closer in skill, but the rewards wouldn't have to change. (I think they will, but they wouldn't have to.) Nor would the game have to shift towards PtW. And there's no way they would do player skill ONLY. Obviously, you'd only ever want to run your best hangar if they did that, like you mentioned. (They might as well get rid of most of he bots if they went that route.) I'm sure there will be a significant MM penalty, though, on skilled players running light bots, cutting down on clubbing.
|
|
|
Post by Strayed on Dec 8, 2016 10:57:07 GMT -5
BTW... that game is impressive Strayed. Thanks!
|
|
|
Post by SlowReflexes on Dec 8, 2016 10:58:13 GMT -5
In our current system it more or less segregates people by matching like equipment against like, and the measure of how good you are is your win %. The new system will explicitly adjust matches so that everyone approaches a 50% win rate. In this case, the measure of how good you are will be your ranking, which will be some kind of representation of your strength as a player. That's nice for bragging rights, but doesn't fit *at all* with the current in game monetary reward system. As the game currently stands this will explicitly make this game PtW. Not at all. You're going to have stiffer competition for that Au if your teammates are closer in skill, but the rewards wouldn't have to change. (I think they will, but they wouldn't have to.) Nor would the game have to shift towards PtW. And there's no way they would do player skill ONLY. Obviously, you'd only ever want to run your best hangar if they did that, like you mentioned. (They might as well get rid of most of he bots if they went that route.) I'm sure there will be a significant MM penalty, though, on skilled players running light bots, cutting down on clubbing. Yes they would have to change the reward system. With MM designed to give everyone a 50% win rate, the current reward system would be like participation trophies. If you want to do better than 'you showed up, you have a 50% chance of a trophy' it's PtW time.
|
|
|
Post by Muhlakai on Dec 8, 2016 11:02:25 GMT -5
Yes they would have to change the reward system. With MM designed to give everyone a 50% win rate, the current reward system would be like participation trophies. If you want to do better than 'you showed up, you have a 50% chance of a trophy' it's PtW time. People with the same win% don't typically all deal the same damage, grab the same number of beacons, etc. Your assumption is that everyone at the same level plays exactly the same way, which just isn't the case.
|
|
|
Post by Strayed on Dec 8, 2016 11:02:37 GMT -5
Possibly, old reward system(10 gold for highest damage, etc. and old silver reward system) for winners and the current reward system for the losers?
|
|
|
Post by Divine Thunder on Dec 8, 2016 11:31:54 GMT -5
Now, i have not read every post in detail, so pardon if i do not really grasp what is going on. However, i believe many have jump to conclusions concerning the up coming matchmaking. First, ELO was not expressly memtioned. Yes, ELO elements may be implimented but from I gleened from the discussions, elemts from other systems are also being incorporated. If it was going to be a pure ELO, I would thing we would already be operating under it.
I predict, i will have to build a brand new hanger after the new mm is introduced. I am prepared to do that. Thus, i suggest, community, save the belly aching and celebrating until we can actually give it a spin around the block. What we get will not please everyone and may surprise you.
|
|
|
Post by ShutUpAndSmokeMyWeed on Dec 8, 2016 11:33:24 GMT -5
First of all, the new system is not going to be the polar opposite of the current one. It's simply going to use skill as a major factor in calculating matchmaker rating. Hangar power will still likely play a significant role in matchmaking, among other things. I take partial blame for this as I should have made it more clear in the OP.
Keep in mind that not even Pixo has finalized the new system yet, because as you guys have mentioned there are many potential problems that need to be solved before release. I don't want to sound like a Pixo muppet but I can assure you that they've already considered all of the issues brought up here (and more) and put a lot of thought into coming up with ways to address them. They obviously won't tell anyone the details of the system, but by the time it's released they could have added so many things on top of the core MMR concept that we'd hardly recognize it.
People from Pixonic do occasionally browse these forums, so if you want to do something positive for the game then suggesting ways to fix problems with the new system is far more valuable than just complaining about the same things over and over.
|
|
|
Post by ł⸰§ĦȺĐ◎ŴƧŦḀɌ on Dec 8, 2016 11:38:55 GMT -5
Pixonic is a for-profit company. Bear that in mind any time you try to apply 'logic' to their actions. It's a balance of factors that creates a popular AND profitable mobile game. If they eliminate all P2W vectors, they go broke because everybody will become freeloaders even with high popularity. If they go all out and make everything about P2W, they will lose popularity and thus most players except so-called whales, and go broke. Most of the dire predictions in this thread are making a leap of faith to one extreme or another. Pixonic has been at this for a long time, they aren't stupid, and they aren't going break the game (it's their cash-cow after all) trying to please everybody. They need some P2W features, and I'm fine with that as long as there is a balance with F2P features.
The gold farming/seal clubbing issue, however, does need to be addressed. I'm sure they've noticed a significant drop-off in account activity when pilots hits the current 'silver' tier and the swarms of TM-Geps and now Aphid-Geps. People give up and move on and that's a loss of potential revenue for the company. This point has been re-hashed so many times it's starting to turn into background noise. Now I know I've gotten myself a bit of a reputation for advocating farming/clubbing but that's not entirely true. I'm an advocate of taking all advantages you can find in a system, seal clubbing just happens to be available here. The full reality is that I FOR anything Pixonic wants to try to eliminate the potential for clubbing.
|
|